October 1, 2006;
Functional analysis of Sox8 during neural crest development in Xenopus.
Among the families of transcription factors expressed at the neural plate border
, Sox proteins have been shown to regulate multiple aspects of neural crest
, exhibit overlapping expression domains in neural crest
progenitors, and studies in mouse suggest that Sox8
functions redundantly with Sox9
during neural crest
development. Here, we show that in Xenopus, Sox8
accumulates at the lateral
edges of the neural plate
at the mid-gastrula stage
; in contrast to its mouse and chick orthologs, Sox8
expression precedes that of Sox9
in neural crest
progenitors. Later in development, Sox8
expression persists in migrating cranial crest cells as they populate the pharyngeal arches
and in trunk neural crest
cells, in a pattern that recapitulates both Sox9
expression domains. Although morpholino-mediated knockdown of Sox8
protein did not prevent the formation of neural crest
progenitors, the timing of their induction was severely affected. This delay in neural crest
specification had dramatic consequences on the development of multiple lineages of the neural crest
. We demonstrate that these defects are due to the inability of neural crest
cells to migrate into the periphery, rather than to a deficiency in neural crest
progenitors specification and survival. These results indicate that the control of Sox8
expression at the neural plate border
is a key process in initiating neural crest
formation in Xenopus, and highlight species-specific differences in the relative importance of SoxE proteins during neural crest
[+] show captions
Fig. 1. Sox8 expression in neural crest progenitors and their derivatives. (A) Sox8 is first detected at the gastrula stage in a ventrolateral domain around the blastopore. At stage 11.5 (B-D) Sox8 expression around the blastopore persists and additional expression is detected lateral to the neural plate (arrows). Vegetal (A), lateral (B,C; anterior towards right) and dorsal (D; anterior towards the top) views. Comparison of Sox8 (E) Sox9 (F) and Sox10 (G) expression in sibling stage 14/15 embryos illustrates that all three genes are expressed in the presumptive neural crest. Dorsal views, anterior towards the top. (H) Section of a stage 15 embryo illustrates the expression of Sox8 in both the lateral (arrows) and the medial neural crest (arrowheads). At stage 17 (I,J), Sox8 persists in the neural crest region and is also expressed anterior to the neural plate in the prospective cement gland (arrow). As the neural tube closes (K,L), Sox8 is detected in the migrating neural crest cells in the cranial region (arrows), and the premigratory cells in the trunk neural crest (arrowheads). Dorsal view, anterior towards top (I,K); lateral view, anterior towards the left (L); cranial view (J). (M) Comparison of Sox8, Sox9 and Sox10 expression at the tailbud stages. Dorsal views of stage 25 embryos, anterior towards the right. Sox8, Sox9 and Sox10 are co-expressed in the migrating cranial neural crest. Posteriorly, although Sox8 and Sox10 are both expressed in trunk neural crest cells, Sox9 is downregulated in this cell population. At stage 35 (left panels, lateral views), Sox8 is detected in the cranial neural crest similar to Sox9; however, at this stage, Sox10 starts to be downregulated in the branchial arches. Sox8, Sox9 and Sox10 are co-expressed in the otic vesicle at this stage (arrows). Ventral views (right panels) showing colocalization of Sox8 and Sox9 in the pancreatic rudiments (arrowheads) where Sox10 is not detected.
Fig. 2. Comparison of the onset expression of Sox8 with other neural plate border-specific genes. (A) Summary of the onset of expression of Sox8 and seven other neural plate border-specific genes in Xenopus. The developmental stages are according to Nieuwkoop and Faber (Nieuwkoop and Faber, 1956). (B) Developmental expression of Pax3, Snail, Myc, Slug and Sox10 from stage 11 to stage 14 by whole-mount in situ hybridization. Dorsal views, anterior towards the top.
Fig. 3. Sox8-depleted embryos fail to express Sox10 at the neural plate border. (A) Increasing amounts of Sox8mo (10 ng, 100 ng and 1000 ng) blocks translation directed by Sox8 mRNA. The same morpholino (500 ng) fails to block Sox9 and Sox10 translation. (B) Embryos injected in one blastomere at the two-cell stage with 30 ng of Sox8mo exhibit reduced Sox10 expression at stage 17, while Slug, Sox9 and Sox2 expression appears largely unaffected at this stage. (C) Quantification of the in situ hybridization results. The numbers in parenthesis indicate the number of embryos analyzed. (D) Sox8mo (500 ng) does not interfere with translation of a mutated Sox8 mRNA (mSox8). (E) Rescue experiments were performed by injection of an animal dorsal blastomere at the eight-cell stage. Bilateral Sox10 expression is rescued in Sox8mo-injected embryo by co-injection of mSox8 mRNA (Sox8mo+mSox8). Single injection of mSox8 expanded the Sox10 expression domain. Injection of a 5 bp mis-matched Sox8 morpholino (Sox8mis) had no effect on Sox10 expression. (F) Quantification of the in situ hybridization results. The numbers in parentheses indicate the number of embryos analyzed. (B,E) Dorsal view, anterior is towards the top. RNA encoding the lineage tracer β-galactosidase was co-injected to identify the injected side (red staining) (left side in B and right side in E).
Fig. 4. Sox8 regulates the onset of expression of Slug and Sox9. (A) Embryos injected in one blastomere at the two-cell stage with 30 ng of Sox8mo exhibited reduced Sox9 and Slug expression at stage 12.5 and stage 14. Lateral view in all panels, anterior is towards the right (control side) or to the left (injected side). (B) Quantification of the in situ hybridization results. The numbers in parentheses indicate the number of embryos analyzed.
Fig. 5. SoxE factors function redundantly at the neural plate border. (A) Sox8, Sox9 or Sox10 can equally rescue Slug expression levels at stage 14 and expand Sox10 expression domain in Sox8-depleted embryos. In these experiments, Sox10 expression was evaluated using a probe against Sox10 3′ UTR. (B) Western blot analysis. Detection of SoxE-GR proteins in extracts from injected embryos collected at stage 17 after injection at the two-cell stage. The fusion proteins are expressed at similar levels, as revealed with α-GR antibody. -, uninjected control embryo;α -tubulin is presented as a loading control.
Sox8 (SRY (sex determining region Y)-box 8) gene expression in Xenopus tropicalis embryo, assayed via in situ hybridization, NF stage 25, lateral view, anterior left, dorsal up.
Fig. 8. Sox8mo-injected embryos have defects in neural crest migration. (A) Phosphohistone H3 immunostaining (α-pH3) shows no difference in the pattern of cell proliferation in stage 17 embryos that received unilateral injection of Sox8mo (left side; FITC label) when compared with the uninjected side. (B) TUNEL staining shows no difference in the pattern of cell death in stage 17 embryos that received unilateral injection of Sox8mo (left side; FITC label) when compared with the uninjected side. In A,B, embryos are viewed from the dorsal side, anterior towards the top. (C) The migration of cranial neural crest cells into the pharyngeal arches visualized by Sox9 and Sox10 expression is severely perturbed in Sox8-depleted embryos (brackets). These cells appear to accumulate lateral to the hindbrain. RNA encoding the lineage tracer β-galactosidase was co-injected to identify the injected side (red staining). Embryos are viewed from the lateral side, anterior towards the left (left panels) or anterior towards the right (right panels). (D) Tissue section of Sox8mo-injected embryos showing accumulation of Sox10-positive cells (small bracket; outlined in red) lateral to the hindbrain on the injected side (arrow); on the control side, Sox10-positive cells have initiated their migration (large bracket; outlined in red). The black outline indicates the position of the hindbrain (hb). (E) TUNEL staining of a stage 25 embryo that received unilateral injection of Sox8mo (left panel; dorsal view, anterior to top). The injected side (arrow) is characterized by reduced pharyngeal arches. Higher power views of the cranial regions on the injected side show no significant increase in TUNEL-positive cells (right panel; lateral view, anterior towards left, dorsal towards top) when compared with the uninjected side (middle panel; lateral view, anterior towards the right, dorsal towards the top).